Submissions from pytorch.org

		From Minutes to Seconds: LLM-Guided Autotuning for Helion Kernels (pytorch.org)
		3 points by matt_d 12 hours ago \| past \| discuss
		PyTorch's playbook for AI coding, as of May 2026 (pytorch.org)
		3 points by matt_d 17 days ago \| past
		When does fragmentation occur in the CUDA caching allocator? (pytorch.org)
		15 points by matt_d 17 days ago \| past \| 1 comment
		PyTorch 2.12 Release (pytorch.org)
		7 points by gmays 29 days ago \| past
		Running PyTorch Models on Apple Silicon GPUs with the ExecuTorch MLX Delegate (pytorch.org)
		2 points by tosh 30 days ago \| past
		Running PyTorch Models on Apple Silicon GPUs with the ExecuTorch MLX Delegate (pytorch.org)
		2 points by 0bytematt 31 days ago \| past
		PyTorch 2.12 Released (pytorch.org)
		1 point by 0bytematt 36 days ago \| past
		In-Kernel Broadcast Optimization: Co-Designing Kernels for RecSys Inference (pytorch.org)
		2 points by gmays 37 days ago \| past
		PyTorch DevLog (pytorch.org)
		2 points by matt_d 39 days ago \| past
		SMG: The Case for Disaggregating CPU from GPU in LLM Serving (pytorch.org)
		3 points by gmays 44 days ago \| past
		AutoSP from PyTorch (pytorch.org)
		1 point by gmays 44 days ago \| past
		AutoSP: Long-Context LLM Training via Compiler-Based Sequence Parallelism (pytorch.org)
		1 point by matt_d 50 days ago \| past
		A Primer on LLM Post-Training (pytorch.org)
		2 points by hyperpape 51 days ago \| past
		Optimizing Effective Training Time for Meta's Recommendation/Ranking Workloads (pytorch.org)
		1 point by gmays 57 days ago \| past
		Monarch: An API to Your Supercomputer (pytorch.org)
		1 point by gmays 69 days ago \| past
		SOTA Normalization Performance with Torch.compile (pytorch.org)
		1 point by salkahfi 72 days ago \| past
		PyTorch 2.11 Released (pytorch.org)
		7 points by 0bytematt 87 days ago \| past
		TorchSpec: Speculative Decoding Training at Scale (pytorch.org)
		2 points by zagwdt 88 days ago \| past
		Generalized Dot-Product Attention: Tackling Real-World Challenges in GPU Kernels (pytorch.org)
		1 point by matt_d 3 months ago \| past
		PyTorch Broadcasting Semantics (pytorch.org)
		1 point by tosh 3 months ago \| past
		Mooncake Joins PyTorch Ecosystem (pytorch.org)
		1 point by mji 4 months ago \| past
		PyTorch Now Uses Pyrefly for Type Checking (pytorch.org)
		5 points by ocamoss 4 months ago \| past
		Building Highly Efficient Inference System for Recommenders Using PyTorch (pytorch.org)
		2 points by mfiguiere 4 months ago \| past
		Warp Specialization in Triton: Design and Roadmap (pytorch.org)
		2 points by matt_d 5 months ago \| past
		NeuralOperator Joins the PyTorch Ecosystem (pytorch.org)
		3 points by williamjsdavis 6 months ago \| past
		KernelFalcon: Autonomous GPU Kernel Generation via Deep Agents (pytorch.org)
		3 points by sadiq 7 months ago \| past
		DeepInverse Joins the PyTorch Ecosystem (pytorch.org)
		2 points by jeremyscanvic 7 months ago \| past
		Helion: A high-level DSL for performant and portable ML kernels (pytorch.org)
		150 points by jarbus 7 months ago \| past \| 49 comments
		Torchforge – a PyTorch native library for scalable RL post-training (pytorch.org)
		2 points by Palmik 7 months ago \| past
		ExecuTorch 1.0 (pytorch.org)
		2 points by jonbaer 7 months ago \| past
		More